Application Of Histogram Of Apparent Diffusion Coefficient In The Grading Of Hepatocellular Carcinoma

Objective:To explore the application value of the apparent diffusion coefficient(ADC)histogram parameters in the pathological differentiation of hepatocellular carcinoma(HCC).Methods:A retrospective analysis of 47 cases of HCC confirmed by pathologyof which 20 cases were highly differentiated,15 cases were moderately differentiated,and 12 cases were poorly differentiated.All patients underwent routine MR unenhanced scanning,enhancedand diffusion-weighted imaging sequence,post-process the DWI image on the GE AW4.6 workstation and obtain the ADC image,refer to the horizontal axis T2WI image of the same plane,delineate the region of interest(ROI)layer by layer,and extract the image through MATLAB software Information,generate ADC histograms based on the entire tumor area,and obtain the histogram average(ADC＿mean)、minimum(ADC＿min)、maximum(ADC＿max)、ADC＿10th,20th,30th,40th,50th,60th,70th,80th,90th percentiles、Quarters(ADC＿25th,ADC＿75th)、kurtosis、skewness.Use ICC to test the consistency of the measurement results of two observers,compare the differences of various parameters in high,medium,and low HCC,and use receiver operating characteristic(ROC)curve to evaluate the diagnostics of histogram parameters in identifying HCC with different levels of differentiation efficacy.Results:1.The measurement results of all parameters by the two observers were consistent.2.ADC values in the ADC histogram parameters increased with the degree of differentiation of HCC.The parameters were divided into high differentiation group,non-high differentiation group,low differentiation group and non-low differentiation group for analysis,and the differences among parameters ADC＿min,ADC＿30th,ADC＿70th and ADC＿75th in the high differentiation group and non-high differentiation group were statistically significant.The differences between the parameters ADC＿min,ADC＿10th and ADC＿60th of the poorly differentiated group and the non-poorly differentiated group are statistically significant.3.AUC values of ADC＿min,ADC＿30th,ADC＿70th and ADC＿75th are0.861,0.806,0.833 and 0.889,respectively,in differentiating highly differentiated group from non-highly differentiated group,and ADC＿75th has the highest diagnostic efficiency with a cutoff value of 1371×10-6mm~2/s.The sensitivity and specificity were 87.3%and 67.6%,respectively.In distinguishing the poorly differentiated group from the non-poorly differentiated group,the AUC values of ADC＿min,ADC＿10th and ADC＿60th are 0.917,0.847 and 0.878,respectively.Among them,ADC＿min has the highest diagnostic efficiency with 388×10-6mm~2/s as the truncation value.The sensitivity and specificity were 91.7%and 77.8%,respectively.Conclusion:ADC histogram has certain application value in evaluating the pathological grading of hepatocellular carcinoma,and the histogram parameters are correlated with the differentiation degree of HCC.The diagnostic efficiency of ADC＿75th was the highest in differentiating highly differentiated group from non-highly differentiated group.The diagnostic efficiency of ADC＿min was the highest in differentiating the poorly differentiated group from the non-poorly differentiated group.